Constitutional mismatch repair deficiency mimicking Lynch syndrome is associated with hypomorphic mismatch repair gene variants

Abstract

Lynch syndrome (LS) and constitutional mismatch repair deficiency (CMMRD) are distinct cancer syndromes caused, respectively, by mono- and bi-allelic germline mismatch repair (MMR) variants. LS predisposes to mainly gastrointestinal and genitourinary cancers in adulthood. CMMRD predisposes to brain, haematological, and LS-spectrum cancers from childhood. Two suspected LS patients with first cancer diagnosis aged 27 or 38 years were found to be homozygous for an MMR (likely) pathogenic variant, MSH6 c.3226C>T (p.(Arg1076Cys)), or variant of uncertain significance (VUS), MLH1 c.306G>A (p.(Glu102=)). MLH1 c.306G>A was shown to cause leaky exon 3 skipping. The apparent genotype-phenotype conflict was resolved by detection of constitutional microsatellite instability in both patients, a hallmark feature of CMMRD. A hypomorphic effect of these and other variants found in additional late onset CMMRD cases, identified by literature review, likely explains a LS-like phenotype. CMMRD testing in carriers of compound heterozygous or homozygous MMR VUS may find similar cases and novel hypomorphic variants. Individualised management of mono- and bi-allelic carriers of hypomorphic MMR variants is needed until we better characterise the associated phenotypes.

Fig. 1. Clinical course of Case 1.

The patient’s timeline of clinical presentation (red boxes), diagnosis (green boxes), and management (blue boxes). The scale bar in the bottom right corner of each immunohistochemistry image represents 100 μm. Immunohistochemistry controls to demonstrate specificity of staining are available in Supplementary Fig. S1.

Fig. 2. Case 1 patient’s pedigree.

The circles denote females and the squares denote males. Filled symbols indicate the individual had a tumour, with the age at tumour diagnosis being given in years (y) if known. Deceased individuals are indicated with a strike-through. The patient is denoted by the arrow.

Fig. 3. Clinical course of Case 2.

The patient’s timeline of clinical presentation (red boxes), diagnosis (green boxes), and management (blue and yellow boxes). The scale bar in the bottom right corner of each immunohistochemistry or haematoxylin and eosin (H&E) image represents 100 μm. Immunohistochemistry controls to demonstrate specificity of staining are available in Supplementary Fig. S1.

Fig. 4. Case 2 patient’s pedigree and confirmation of a splice effect of MLH1 c.306G>A.

a The patient’s pedigree. The circles denote females and the squares denote males. Filled symbols indicate the individual had a tumour, with the age at tumour diagnosis being given in years (y) if known. Deceased individuals are indicated with a strike-through. The patient is denoted by the arrow. b Transcript analysis by PCR amplification of cDNA generated from RNA extracted from PHA-stimulated and puromycin-treated short-term cultures of peripheral blood leukocytes. The PCR primers used are located in MLH1 exon 1 and exon 4. The expected amplicon sizes are 333 bp from full-length MLH1 transcript and 234 bp from MLH1 transcript lacking exon 3 (99 bp). C1-C3: control (MLH1 wild type) leukocyte cDNA. P: patient (MLH1 c.306G>A homozygous) leukocyte cDNA. Ø: negative-template control. c Analysis of minigene transcripts by PCR amplification of cDNA from RTB minigene-transfected cell lines (HEC-155, HEC-1B, HRT-18, and HEK-293) using primers located in RTB minigene exon 1 and exon 4 (Supplementary Figure S7). The expected amplicon sizes are 267 bp from RTB minigene transcript containing MLH1 exon 3 and 168 bp from transcript lacking MLH1 exon 3 (99 bp). WT: cDNA amplicons from cells transfected with RTB minigene containing wild type MLH1 exon 3. V: cDNA amplicons from cells transfected with RTB minigene containing variant (c.306G>A) MLH1 exon 3. Ø: negative-template control. Note: Amplification of cDNA from the RTB minigene containing variant c.306G>A MLH1 exon 3 produces an unexpected third, larger amplicon that appears to be a heteroduplex of the other products based on Sanger sequencing data (Supplementary Fig. S5).